Digital Marketing

Measuring Digital Marketing ROI When Half Your Sales Close on the Phone

By Rajeev Gupta
ROI in digital marketing when sales close on the phone: Google, Meta, YouTube and Instagram ads on a laptop lead to an incoming call, a smiling customer on the phone, and rising stacks of gold coins under an ROI arrow

Here's a scene we see often. A business spends ₹60,000 a month on Google Ads and SEO. The dashboard shows 40 form fills. The owner knows that most of the people who actually bought called a number or sent a WhatsApp message first. So when the agency says "ROI is 80%", the owner shrugs, because the number came from a system that can't see how the business really sells.

This guide is about ROI in digital marketing for that kind of business. Think of a clinic, a factory outlet, a coaching centre, a B2B supplier or a D2C brand whose buyers ask for a price on WhatsApp before ordering. The standard guides from big software vendors assume an online checkout and a CRM. Most Indian small and mid-sized businesses have neither. We'll show you how to measure return anyway, with tools you probably already own.

The short version: ROI is simple arithmetic. The hard part is getting a trustworthy revenue number to put into it. Fix the revenue side first and everything else gets easier.

The ROI formula, and why it looks wrong for most SMBs

The textbook formula is the same one used for any investment. According to the general definition of return on investment on Wikipedia, it's net gain divided by cost. For marketing:

ROI = (revenue from marketing − marketing cost) ÷ marketing cost × 100

Say you spend ₹1,00,000 in a month and it brings ₹3,00,000 in sales. Net gain is ₹2,00,000, so ROI is 200%. You earned ₹2 for every ₹1 spent, on top of getting the ₹1 back.

Two things trip people up straight away.

First, revenue isn't profit. If your gross margin is 30%, that ₹3,00,000 of sales left only ₹90,000 to cover the ₹1,00,000 you spent. On a margin basis you lost money. Big-brand guides often skip this, because software and subscription businesses have margins above 70%. A trading business, a clinic or a manufacturer doesn't. Use gross profit as the numerator whenever you can, and call that number "profit ROI" so nobody confuses the two.

Second, the cost side is bigger than the ad bill. Count agency fees, freelancer fees, tool subscriptions, landing page work, call tracking, and the hours your own staff spend answering leads. Leaving these out makes ROI look better than it is, and the gap tends to show up later as a surprise.

Now the part that really breaks the formula for SMBs. Revenue comes from the sales system, and ad platforms only see what happens on the website. If the customer clicked your ad, read your page, then phoned you and paid at your shop three days later, the platform recorded a click and nothing else. As far as the dashboard is concerned, that ad produced zero revenue. We've seen this make a profitable campaign look like a waste, and the owner switched it off.

Where attribution breaks: calls, WhatsApp, walk-ins

Attribution just means answering "which marketing effort caused this sale?" It breaks in four common places for Indian SMBs.

Phone calls. Someone sees your ad on their phone and taps the call button, or copies the number from your site and dials it later. Nothing on your website records that. Your Google Analytics 4 property shows a visit and the visitor leaves.

WhatsApp. A floating WhatsApp button is a very useful addition to to an Indian business site, and also one of the least measured. The click opens the WhatsApp app. The conversation then happens entirely outside any tracking.

Walk-ins and offline closes. The enquiry came online, but the deal was signed in an office or paid in cash. The marketing system never hears about it.

Long gaps. A buyer researches in October, asks for a quote in November, and pays in January. A 30-day report won't connect those dots unless you record the original source somewhere.

Our own team has run into all four on client accounts. In our experience, the first reaction is to blame the channel, usually SEO, because SEO feels slow and is hard to see. The better first reaction is to ask what percentage of sales the dashboard can actually see. If the answer is "less than half", the ROI figure you're staring at is a floor, not the real number.

You can estimate that percentage quickly. Take your last 20 paying customers and ask each one, or look at your own notes, how they first found you and how they got in touch. Tally it. Even a sample of 20 customers tells you whether you have a tracking problem or a marketing problem, and those need very different fixes.

Tracking calls and WhatsApp clicks

The goal is simple: turn every call and every WhatsApp tap into a recorded event with a source attached. Here's the practical way to do it, from cheapest to most complete.

Step 1: Record WhatsApp and call-button clicks as events

In Google Tag Manager, set up a click trigger for links that start with https://wa.me/ or tel:. Send each one to GA4 as a named event, for example whatsapp_click and call_click. Then mark both as key events. Google's Analytics Help page on key events explains that key events are the actions you flag as especially important to your business, and that's what lets them show up in acquisition reports tied to a source.

This takes an afternoon. It records the intent to talk, not the conversation itself, so treat it as a lead signal rather than a sale.

Step 2: Use a tracked number for ads

If you run Google Ads, you can show a Google forwarding number on your site and in ads, so that calls from that traffic are attributed to the campaign. Google's help page on tracking calls to a phone number on a website walks through the setup. You can also set a minimum call length, so a 5-second wrong number doesn't count as a conversion. We usually start with 60 seconds for service businesses and adjust once we've heard a sample of recordings.

For SEO and organic traffic, a call-tracking tool that swaps the displayed number by source works the same way. Pick one that stores the landing page and the keyword or campaign with every call.

A note on privacy before you record anything. If you plan to record calls or store customer phone numbers against marketing sources, tell callers and keep the data only as long as you need it. India's Digital Personal Data Protection Act, 2023 sets rules on consent and purpose for personal data, and the Ministry of Electronics and Information Technology keeps the official material on its data protection framework page. Ask your lawyer how it applies to your business. A one-line notice at the start of the call is the usual starting point.

Step 3: Make WhatsApp conversations traceable

This is where most businesses stall. A few approaches that work, in rising order of effort:

  • Put a different pre-filled message on each landing page's WhatsApp link. For example, the Google Ads page says "Hi, I saw your ad for dental implants" and the SEO blog says "Hi, I read your guide on implant costs". Your staff can then see the source in the first message, without asking.
  • Give each campaign its own WhatsApp number if volume justifies it. This is clunky, but it's clean.
  • Use the WhatsApp Business API through a provider that supports click-to-WhatsApp ads and passes the source into a simple inbox or CRM.

We've found the pre-filled message approach gets about 80% of the benefit for no cost. It also nudges the customer to say what they want, which helps your team reply faster.

Step 4: Add one question to the call or chat

Train whoever answers to ask, "How did you hear about us?" and log the answer in a shared sheet. It feels low-tech, and it is. It's also the only method that catches word of mouth, a printed ad, and the person who saw your Instagram post and then Googled your name. Treat the answers as a cross-check on your software numbers.

Importing offline sales back into Google and Meta

Tracking the call gets you to the lead. The next step is the one almost nobody in the SMB space does: telling the ad platforms which leads became paying customers.

According to Google's help page on offline conversion imports, you can upload conversions that happened after a click, such as a sale closed by phone, and Google matches them back to the ad click using an identifier called the GCLID. Once Google sees which clicks led to real sales, its bidding can aim for more of that kind of click, not just more form fills.

Here's how this works for a business with no CRM.

  1. Capture the click ID. Add hidden fields to your enquiry form that store the GCLID from the landing page URL. For WhatsApp and calls, you can capture it when the visitor first lands and attach it to the lead in your sheet through the form or a tracking tool.
  2. Keep a lead sheet. One row per lead, with date, source, GCLID where you have it, contact name, status and sale value. Google Sheets is fine.
  3. Update the status when a deal closes. This takes your sales person 30 seconds per sale. It's the discipline that makes everything else work.
  4. Upload weekly. Export the closed rows with GCLID, conversion time and value, and upload them to Google Ads as a CSV. A scheduled sheet import can automate this later.

Meta has a similar idea. Offline events and Conversions API let you send purchase or qualified-lead events back to your ad account, matched by hashed email or phone number. The setup is more technical than Google's, so most small businesses should start with Google and add Meta once the lead sheet habit is solid.

There's a trap here. Don't import every enquiry as a conversion just to make the dashboard look better. If a lead was a student asking about a job, or a wrong number, it shouldn't count. We prefer two conversion actions: "qualified lead" (a real person who fits what you sell) and "closed sale" (with value). Bidding on closed sales is ideal, but with fewer than about 30 sales a month the platform may not get enough signal, so qualified leads work as the bridge.

If your business has long sales cycles, remember that Google only matches imports within a window after the click. Check the current window in the help page when you set up, and upload before it closes.

A one-page monthly ROI sheet

You don't need a dashboard tool. One sheet, updated once a month, is enough to answer the question that matters: did marketing pay for itself?

Build it with one row per channel and these columns:

  • Spend: ad spend + fees + tools allocated to that channel.
  • Leads: total enquiries from calls, forms, WhatsApp and walk-ins.
  • Qualified leads: the ones that fit your offer and budget.
  • Closed sales: count and total value.
  • Gross profit: sales value × your gross margin.
  • Cost per lead: spend ÷ leads.
  • Cost per sale: spend ÷ closed sales.
  • Profit ROI: (gross profit − spend) ÷ spend.

Here's a worked example with made-up numbers, just to show the mechanics. It isn't a client result.

Suppose a clinic spends ₹50,000 in a month on Google Ads, including the agency fee. It receives 90 enquiries: 30 form fills, 40 calls and 20 WhatsApp chats. Of those, 45 are qualified. The front desk closes 15 patients with an average first-month billing of ₹8,000, so sales are ₹1,20,000. With a 50% gross margin, gross profit is ₹60,000.

  • Cost per lead: ₹50,000 ÷ 90 = about ₹556
  • Cost per sale: ₹50,000 ÷ 15 = about ₹3,333
  • Profit ROI: (₹60,000 − ₹50,000) ÷ ₹50,000 = 20%

Now look at what the ad platform alone would have reported: 30 form fills, and no sales at all. The owner would have concluded that the campaign cost ₹1,667 per lead and produced nothing. With calls and WhatsApp counted and sales logged, the same campaign turns out to be mildly profitable in the first month, before counting that these patients may return. That gap is the whole argument for doing this work.

Add a column called "repeat value" if customers come back. A patient who returns for 4 more visits, or a B2B buyer who reorders every quarter, changes the picture a lot. Use a conservative estimate, such as the average of what your last 20 customers actually spent in 12 months, and label it clearly so nobody mistakes it for a forecast.

Review the sheet monthly. Look at it quarterly for trends. A single month is noisy, especially when sales cycles run longer than 30 days.

ROI by channel: SEO vs ads vs WhatsApp

Each channel produces return differently, so comparing them on one month's numbers is unfair. Here's how we think about them.

Paid ads (Google, Meta). Fastest to measure. You pay per click, so cost is clear and results start within days. The risk is that returns stop the moment you stop paying. Judge ads on cost per qualified lead and cost per sale, and review at 4 to 8 weeks after any major change.

SEO. Slower and compounding. You pay up front for content, technical fixes and links, and the benefit builds over 3 to 6 months or more, with no per-click fee. Because SEO traffic often arrives through search for a brand name or a specific question, a lot of it ends in a call or WhatsApp message and gets missed. Our team advises clients to compare SEO on a 6 to 12 month basis. Measure organic leads and sales per month against total monthly SEO cost, then watch the trend rather than any single month. Our pricing page, digital marketing pricing at Nurotech, lists what typical monthly SEO scopes include, so you can set cost against deliverables.

WhatsApp. This isn't a traffic source. It's a conversion channel that sits behind the others. Count it as the place where leads from ads, SEO and social turn into sales, and measure its quality by response time and close rate. A common finding when businesses start tracking is that WhatsApp leads close at a higher rate than form leads, because the person was willing to start a conversation. Check whether that's true for you before you shift budget on that assumption.

Email and social. Both are easy to under-measure. Use tagged links (UTM parameters) on every email and bio link, so GA4 can separate them from direct traffic.

If you're deciding between spending on a team and spending on an agency, our breakdown of in-house, freelancer or agency cost per outcome compares them on cost per lead rather than headline fees, which fits the way this sheet works. And if paid performance is where most of your budget sits, the page on performance marketing in Delhi explains how we structure campaigns around measured outcomes.

ROI versus ROAS

People swap these terms constantly. ROAS, or return on ad spend, is revenue divided by ad spend. A ROAS of 4 means ₹4 of revenue for each ₹1 of ad spend. ROI deducts costs, so it speaks to profit. A campaign can have a healthy ROAS and a negative ROI, if margins are thin or the non-ad costs are large. Use ROAS for quick daily campaign checks and ROI for deciding whether to keep spending.

What a realistic payback period looks like

Payback period is how long it takes for cumulative returns to cover cumulative spend. Nobody can promise one number, and you should be wary of anyone who does. It depends on your sales cycle, your margin, your competition and how well your site converts. What we can give you is the pattern.

  • Paid ads can pay back inside the first 1 to 3 months for businesses with short sales cycles and clear demand, such as local services. For high-ticket or B2B sales with 2-month cycles, expect longer.
  • SEO typically shows early signs (more impressions, more pages indexed) in the first 2 to 3 months, and meaningful lead growth after 4 to 6. Full payback often takes 6 to 12 months. Competitive terms can take longer.
  • Content and email sit in between, and depend heavily on how often you publish and how good the offer is.

The mistake we see most is judging SEO on month 2 numbers and ads on month 12 numbers. Compare like with like, and agree on a review date before you start.

A practical way to set expectations: pick a target payback window with your agency in writing, say 9 months for SEO. Define the metrics you'll check at months 3, 6 and 9, such as qualified leads from organic search, calls from organic pages, and closed sales logged. If the early indicators are flat at month 6, that's the time for a hard conversation and a change of plan, not month 12.

On what counts as a "good" ROI: there isn't a universal figure. A 20% profit ROI on a campaign that scales is better than 300% on one that can only absorb ₹5,000 a month. Compare against your alternatives. If the same money in stock or inventory earns you less, marketing at 20% is fine. What matters is whether the return is positive after margin, repeatable and trending in the right direction.

Mistakes that quietly wreck ROI numbers

These come up in audits again and again.

Counting every lead the same. A ₹500 lead who never buys is more expensive than a ₹2,000 lead who always does. Track cost per sale, not just cost per lead.

Ignoring lead handling. If your team replies to WhatsApp enquiries after 6 hours, a lot of the spend is wasted regardless of how good the campaign is. Before blaming the channel, test your own response time by sending a message from a personal number and timing the reply.

Mixing branded and non-branded traffic. People who search your business name were already going to find you. Separate them in Google Search Console and in your ad accounts, so you can see how much new demand marketing is creating.

Trusting claims you can't verify. If a marketing report quotes a huge return, such as multiples of ad spend, ask what revenue it's based on, over what period, and whether any of it was checked against your own books. A figure you can reconcile with your accounts is worth more than a bigger one you can't.

Changing too many things at once. If you change the ad copy, landing page and bidding in one week, you won't know what moved the result. Change one thing, give it 2 to 4 weeks, then measure.

Where to start this month

You don't have to build the full system at once. A sensible order looks like this:

  1. Week 1: Add click events for WhatsApp and call buttons in GA4, and mark them as key events.
  2. Week 2: Set up the lead sheet with source and status columns, and brief whoever answers calls and chats.
  3. Week 3: Add pre-filled WhatsApp messages per landing page, and a tracked number for ads.
  4. Week 4: Fill in the monthly ROI sheet for the first time, even if it's rough.
  5. Month 2 onwards: Start importing closed sales to Google Ads, and review the sheet monthly.

If you'd like a second pair of eyes on how your tracking currently works, our Discover page explains how we start engagements: with measurement first, then channels. We'd rather show you what your numbers can and can't tell you than hand you a report with a big percentage on it.

Frequently Asked Questions

These are the questions business owners ask us most in the first meeting about marketing returns.

FAQs

Frequently asked questions

How do you calculate ROI in digital marketing?
Subtract your total marketing cost from the revenue it brought in, divide by the cost, and multiply by 100. For a clearer picture use gross profit instead of revenue, and count agency fees, tools and staff time as costs, not just the ad bill. If you spend ₹1,00,000 and earn ₹3,00,000 in sales, ROI on revenue is 200%.
What is a good ROI for digital marketing?
There is no single figure. What matters is that return is positive after your margin, repeatable and improving over time. A 20% profit ROI on a campaign that can scale is often better than 300% on one that can only absorb a small budget. Compare against what the same money would earn elsewhere in your business.
How do I track leads that call instead of filling a form?
Use a tracked phone number for ads and a call-tracking tool for organic traffic, and set GA4 events for taps on call buttons. Add a minimum call length so wrong numbers do not count. Then log each lead in a sheet with its source so you can match calls to closed sales.
How long does SEO take to show ROI?
Early signs such as more impressions and indexed pages often appear in 2 to 3 months, and lead growth usually follows after 4 to 6. Full payback commonly takes 6 to 12 months and longer for competitive terms. Agree on review points at months 3, 6 and 9 and measure organic leads and closed sales at each.
What is the difference between ROI and ROAS?
ROAS is revenue divided by ad spend, so a ROAS of 4 means ₹4 of revenue per ₹1 of ads. ROI deducts all costs and speaks to profit. A campaign can have a good ROAS and still lose money if margins are thin or fees and tools are large. Use ROAS for quick campaign checks and ROI for budget decisions.
Can I measure ROI without a CRM?
Yes. A Google Sheet with one row per lead, covering date, source, status and sale value, is enough to start. Update the status when a deal closes, and upload closed sales to Google Ads weekly. A CRM makes this easier later, but the habit of logging sources and outcomes matters more than the tool.

More from the Nurotech blog

Ready to grow your business?

Fill out the form or call us — let’s find where your next growth comes from.

9659 634 634 [email protected]